A method and system for ductile die cavity optimization based on infrared calibration

Abstract

This invention discloses a method and system for optimizing duct cavity casting based on infrared calibration, relating to the fields of mold design and die-casting process control. This method involves infrared scanning of the plaster mold's outer contour reference surface to obtain the spatial positions of multiple measurement points. These positions are then compared point-by-point with the designed positions to obtain the outer contour offset. Combining the local wall thickness values ​​and outer contour offsets at each measurement point, a heterogeneous shrinkage field inverse distance weighted interpolation algorithm is used to calculate the shrinkage offset of the corresponding points on the inner cavity blade surface, restoring its current spatial position. The current spatial position is then compared with the designed target position to generate offset data. This invention combines the measured outer contour offset with the local wall thickness, using a heterogeneous shrinkage field interpolation algorithm to back-calculate the actual position of the inner cavity blade surface. This allows for the quantitative acquisition of non-uniform shrinkage within the plaster mold before casting. Combined with closed-loop pose adjustment, the offset is corrected to within tolerance, controlling blade wall thickness deviation from its source.

Description

Technical Field

[0001] This invention relates to the field of mold design and die casting process control technology, specifically to a casting method and system for optimizing duct cavity based on infrared calibration. Background Technology

[0002] As high-performance aerodynamic components, the uniformity of the wall thickness of the blades and housing of ducted fans has a significant impact on aerodynamic efficiency and stability. Traditional single-cavity die casting processes typically rely on the outline of plaster molds or metal molds for positioning and calibration, forming the blade and housing structure through integral molds or parting molds. During the casting process, the temperature distribution of the mold, the shrinkage characteristics of the plaster mold, and the complexity of the blade surface all affect the wall thickness uniformity and surface quality of the final product.

[0003] In recent years, with the development of infrared temperature measurement, infrared positioning and online monitoring technologies, applying infrared methods to mold positioning, temperature control and wall thickness detection has become an effective way to improve the die casting accuracy and production consistency of ducted fans.

[0004] The limitations of existing technologies include at least the following problems: When using an integral plaster mold to eliminate blade parting seams, the mold cavity positioning calibration mainly relies on the outer contour reference of the plaster mold, making it difficult to simultaneously acquire the actual spatial position of the inner cavity blade surface. The calibration process usually assumes that the inner and outer structures of the plaster mold shrink uniformly and in the same direction. In actual molding, local wall thickness differences and individual mold differences may cause the shrinkage distribution to deviate from the preset value, resulting in a hidden eccentricity where the external reference is aligned but the inner cavity blade surface is offset. Conventional tooling inspection based on the external reference is difficult to detect this type of offset. Hidden deviations may cause uneven blade wall thickness and accumulate and amplify in continuous batch processing, thereby affecting the smooth and coordinated aerodynamic operation of the ducted fan. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for optimizing duct cavity casting based on infrared calibration, which solves the problem that the positioning of the outer contour of the cavity is difficult to reflect the eccentricity of the inner blades and the uneven wall thickness in existing technologies.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a duct cavity optimization casting method based on infrared calibration, comprising the following steps: performing infrared scanning on the outer contour reference surface of the plaster mold to acquire the actual spatial positions of multiple measurement points; comparing the actual spatial positions of the multiple measurement points with the designed outer contour positions of the plaster mold point by point to obtain the outer contour offset at each measurement point; combining the local wall thickness value and outer contour offset at each measurement point, using a heterogeneous shrinkage field inverse distance weighted interpolation algorithm to calculate the shrinkage offset of the corresponding point on the inner cavity blade surface, and obtaining the current spatial position of the inner cavity blade surface; comparing the current spatial position of the inner cavity blade surface with the design target position to generate inner cavity blade surface offset data; adjusting the clamping position of the plaster mold relative to the casting base plate according to the inner cavity blade surface offset data, including the pose direction and displacement, until the inner cavity blade surface offset detected after adjustment meets the preset tolerance; maintaining the adjusted pose of the plaster mold and performing the duct cavity casting forming operation.

[0007] Furthermore, combining the local wall thickness and outer profile offset at each measurement point, the shrinkage offset of the corresponding point on the inner blade surface is calculated using a heterogeneous shrinkage field inverse distance weighted interpolation algorithm. The specific steps are as follows: using the local wall thickness at each measurement point as a weighting factor, a spatial weighting function of the heterogeneous shrinkage field is constructed; for any point to be calculated on the inner blade surface, multiple measurement points that are spatially closest to the point to be calculated are selected, and the shrinkage offset of the point to be calculated is calculated by weighting based on the inverse ratio of the spatial distance between each measurement point and the point to be calculated, and the spatial weighting function of the heterogeneous shrinkage field.

[0008] Furthermore, the actual spatial positions of multiple measurement points are compared point by point with the designed positions of the plaster mold outline to obtain the outline offset at each measurement point. The specific steps are as follows: perform surface fitting on the actual spatial positions of multiple measurement points to obtain the measured point cloud surface of the plaster mold outline reference surface; perform spatial registration and alignment between the measured point cloud surface and the designed plaster mold outline surface; calculate the deviation distance of each measurement point on the measured point cloud surface along the normal direction of the designed plaster mold outline surface to obtain the outline offset at each measurement point.

[0009] Further, the local wall thickness value at each measurement point is obtained, including: based on the design model of the plaster mold outer contour reference surface and the design model of the inner cavity blade surface, the vertical distance from the plaster mold outer contour reference surface to the inner cavity blade surface is calculated along the normal direction of the location of each measurement point, and this distance is used as the local wall thickness value at each measurement point; wherein, the local wall thickness value and the outer contour offset at each measurement point are used together to input the heterogeneous shrinkage field inverse distance weighted interpolation algorithm.

[0010] Furthermore, the clamping position of the plaster mold relative to the casting base plate is adjusted based on the offset data of the inner cavity blade surface, including the orientation and displacement, until the offset of the inner cavity blade surface detected after adjustment meets the preset tolerance. The specific steps are as follows: decompose the offset data of the inner cavity blade surface into translational and rotational components; adjust the translational degrees of freedom of the plaster mold; after the overall translational deviation is eliminated, adjust the rotational degrees of freedom of the plaster mold; after the adjustment is completed, perform infrared scanning on the outer contour reference surface of the plaster mold again and recalculate the offset of the inner cavity blade surface. If the offset of the inner cavity blade surface does not meet the preset tolerance, repeat the translational and rotational adjustments.

[0011] Further, determining whether the offset of the inner blade surface meets the preset tolerance includes: dividing the inner blade surface into the leading edge region, the middle section region of the blade body, and the trailing edge region of the blade; setting a leading edge tolerance threshold for the leading edge region, a middle section tolerance threshold for the middle section region of the blade body, and a trailing edge tolerance threshold for the trailing edge region of the blade, wherein the leading edge tolerance threshold is less than the middle section tolerance threshold; when the offset of the leading edge region, the middle section region of the blade body, and the trailing edge region of the blade in the inner blade surface offset data all fall within the corresponding tolerance threshold range, it is determined that the offset of the inner blade surface meets the preset tolerance.

[0012] Furthermore, the shrinkage offset of corresponding points on the inner cavity blade surface is calculated. The specific steps are as follows: Discretize the inner cavity blade surface into grid nodes; select measurement points on the outer contour reference surface of the plaster mold that correspond to each grid node along the plaster mold wall thickness direction; take the outer contour offset at the corresponding measurement point as the initial value, multiply it by the shrinkage attenuation coefficient determined by the local wall thickness value, and calculate the shrinkage offset of each grid node on the inner cavity blade surface node by node.

[0013] Furthermore, before calculating the shrinkage offset of corresponding points on the inner cavity blade surface by combining the local wall thickness values ​​and outer contour offset at each measurement point using the heterogeneous shrinkage field inverse distance weighted interpolation algorithm, the local wall thickness value is corrected. The specific steps are as follows: Before the plaster mold enters the mold closing preparation state, the preheating temperature of the plaster mold is collected; based on the linear expansion coefficient of the plaster mold material and the preheating temperature, the wall thickness expansion of the plaster mold in the preheating state is calculated; the design wall thickness value between the design model of the outer contour reference surface of the plaster mold and the design model of the inner cavity blade surface is superimposed with the wall thickness expansion value to obtain the corrected local wall thickness value.

[0014] Furthermore, the clamping position of the plaster mold relative to the casting base plate is adjusted according to the internal blade surface offset data, including the orientation and displacement. The specific steps are as follows: When the internal blade surface offset data indicates that the internal blade surface is offset to one side, the internal blade surface offset data is converted into a reverse compensation amount for the plaster mold clamping position; the reverse compensation amount is decomposed into at least two progressive adjustment steps according to a preset ratio; after executing the previous progressive adjustment step, the actual spatial position of the plaster mold outer contour reference surface is re-acquired and the internal blade surface offset data is recalculated; then, the subsequent progressive adjustment step is executed according to the recalculated internal blade surface offset data.

[0015] An infrared-calibrated duct cavity optimization casting system includes: an infrared scanning unit for scanning the outer contour reference surface of a plaster mold with infrared light, acquiring the actual spatial positions of multiple measurement points, and also acquiring the preheating temperature of the plaster mold before it enters the mold closing preparation state; an outer contour offset acquisition unit for comparing the actual spatial positions of multiple measurement points with the designed outer contour positions of the plaster mold point by point to obtain the outer contour offset at each measurement point; and a heterogeneous shrinkage field interpolation calculation unit for calculating the wall thickness expansion of the plaster mold in the preheated state based on the linear expansion coefficient of the plaster mold material and the preheating temperature, superimposing the designed wall thickness value with the wall thickness expansion value to obtain the corrected local wall thickness value, and combining the corrected local wall thickness value... The system calculates the shrinkage offset of corresponding points on the inner cavity blade surface using a heterogeneous shrinkage field inverse distance weighted interpolation algorithm, and obtains the current spatial position of the inner cavity blade surface. An inner cavity offset generation unit compares the current spatial position of the inner cavity blade surface with the design target position to generate inner cavity blade surface offset data. A clamping and adjustment unit adjusts the clamping position of the plaster mold relative to the casting base plate based on the inner cavity blade surface offset data, including the pose direction and displacement, until the detected inner cavity blade surface offset after adjustment meets the preset tolerance. A casting execution unit controls the execution of the duct mold cavity casting process under the adjusted pose of the plaster mold.

[0016] The present invention has the following beneficial effects: (1) The infrared calibration-based duct mold cavity optimization casting method obtains the actual spatial position of multiple measurement points on the outer contour reference surface of the plaster mold by infrared scanning, and compares them with the design coordinates point by point to obtain the outer contour offset. Then, the local wall thickness value at each measurement point is introduced as a weighting factor, and the heterogeneous shrinkage field inverse distance weighted interpolation algorithm is executed to calculate the shrinkage offset of the corresponding point of the inner cavity blade surface, thereby restoring the current spatial position of the inner cavity blade surface in the mold preparation stage. This breaks through the inherent limitation of the traditional calibration that the shrinkage inside and outside the plaster mold is uniform and in the same direction. Because the response weight of the region with larger wall thickness in its algorithm is smaller and vice versa, the calculation result can truly reflect the non-uniform shrinkage distribution of the plaster mold caused by the local structural thickness difference. The blade surface offset that was originally closed inside the mold body and difficult to be detected by conventional tooling can be quantitatively revealed before casting, thereby suppressing the transformation of hidden eccentricity into blade wall thickness deviation at the source.

[0017] (2) The infrared calibration-based duct cavity optimization casting method generates offset data by comparing the current spatial position of the inner cavity blade surface with the design target position after obtaining the current spatial position of the inner cavity blade surface. The data is decomposed into translational and rotational components. The decoupled adjustment is carried out in the order of first adjusting the translational degree of freedom to eliminate the overall positional deviation and then adjusting the rotational degree of freedom to eliminate the angular deviation. After each adjustment, the infrared scan and offset calculation are re-executed. If the preset tolerance is still exceeded, the correction is iteratively corrected in a loop, forming a closed-loop calibration process. When converting the offset into clamping position compensation, the total compensation is divided into multiple progressive adjustment steps according to the preset ratio. After each step is completed, the outer contour data is re-acquired and the offset calculation is refreshed to determine the adjustment range of the next step. This effectively prevents overshoot and local stress concentration in the mold caused by a single large adjustment and reduces the risk of secondary deviation caused by human error in one adjustment.

[0018] (3) The infrared calibration-based duct cavity optimization casting method divides the inner cavity blade surface into the blade leading edge area, the blade middle section area and the blade trailing edge area according to the aerodynamic sensitivity, and sets different tolerance thresholds for different areas. The threshold for the leading edge area is stricter than that for the middle section area. Finally, the preset tolerance is only determined when the offset of all areas falls within their respective threshold ranges. As the area where the airflow first contacts and the load is most concentrated, the wall thickness accuracy of the leading edge has a much greater impact on the fan propulsion efficiency and operational stability than other areas. Therefore, it is matched with stricter admission standards. Compared with the conventional approach of using the overall average deviation as the criterion, the partitioned tolerance avoids the hidden danger of local key areas exceeding the tolerance being covered by the overall average value, and ensures that the key aerodynamic parts of each blade can obtain consistent wall thickness guarantee in mass production.

[0019] (4) The infrared calibration-based duct cavity optimization casting system, through the coordinated cooperation of infrared scanning, outer contour offset acquisition, heterogeneous shrinkage field interpolation calculation, inner cavity offset generation, clamping adjustment and casting execution unit, the infrared scanning unit is responsible for acquiring the original spatial point cloud, the outer contour offset acquisition unit completes the registration and deviation extraction, the heterogeneous shrinkage field interpolation calculation unit executes the core algorithm to restore the inner cavity surface position, the inner cavity offset generation unit outputs the deviation data, the clamping adjustment unit drives the translation and rotation degrees of freedom to perform pose correction and triggers re-scanning when the target is not met, and the casting execution unit starts casting after the adjustment is in place. This architecture reduces the uncertainties caused by manual links.

[0020] Of course, any product implementing this invention does not necessarily need to achieve all of the above advantages at the same time. Attached Figure Description

[0021] Figure 1 This is a flowchart of an optimized casting method for duct cavity based on infrared calibration according to the present invention.

[0022] Figure 2 This is a flowchart illustrating the specific steps involved in calculating the shrinkage offset of corresponding points on the curved surface of the inner cavity blade in an infrared-calibrated duct cavity optimization casting method according to the present invention.

[0023] Figure 3 This is a block diagram of an optimized casting system for ducted mold cavities based on infrared calibration, according to the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] The manufacturing of ducted fan housings involves a complete technology chain with multiple coupled processes and progressively increasing precision, encompassing several consecutive stages including casting, process inspection, heat treatment control, dimensional finishing, and final performance verification. In this manufacturing chain, the geometric fidelity of the casting cavity plays a fundamental role: if the physical cavity relied upon by the midstream forming stage cannot accurately reproduce the design intent, a structural disconnect will occur between the upstream digital model and the downstream manufacturing benchmark, causing the forming result to continuously deviate from the theoretical target. Therefore, achieving a paradigm shift from "approximate replication" to "precise pre-tuning" is a core prerequisite for ensuring the casting precision of complex internal cavity components.

[0026] In the aforementioned technology chain, the cavity optimization and pre-adjustment process is a crucial node in transforming the design geometry into a physical cavity. For ducted fan housings using precision casting molds such as plaster molds, the curved surfaces of the inner blades and the flow channel structure are affected by non-uniform temperature fields and material shrinkage characteristics during the casting and cooling process, often causing their actual spatial positions to deviate from the design target. More complexly, the outer contour and inner cavity of such castings do not shrink homogeneously and proportionally; a non-linear mapping relationship exists between the measured outer contour offset and the actual inner cavity offset, determined by wall thickness distribution, structural stiffness, and heat dissipation conditions. The cavity geometry pre-adjustment accuracy achieved in this process directly determines whether the initial physical boundaries faced by subsequent forming, inspection, and finishing stages are consistent with the design model.

[0027] However, existing mold cavity calibration methods suffer from a bottleneck that limits the overall accuracy of the process: typically, only simple dimensional checks are performed on the plaster mold outline, or the overall mold size is scaled based on experience, without establishing a quantitative inversion model between the measured offset of the outline and the non-uniform shrinkage of the inner cavity surface. For ducted fan casings with complex inner cavity blades and varying wall thickness gradients, empirical overall scaling cannot match the differentiated shrinkage patterns of various local areas, resulting in systematic deviations in the positions of key inner cavity surfaces before casting. This pre-adjustment misalignment will trigger serious cascading consequences: the initial deviation of the cavity geometry causes structural misalignment between the inner cavity of the casting after filling and the design datum, making it difficult to distinguish between "cavity deviation" and "filling deformation" in subsequent inspection stages; finishing processes face unexpected allowance distributions; and the uniformity of wall thickness and the consistency of structural function in the final product cannot be guaranteed.

[0028] Please see Figure 1 This invention provides a technical solution: an infrared calibration-based optimized casting method for duct cavity molds, comprising the following steps: infrared scanning of the overall plaster mold outer contour reference surface (e.g., area array infrared scanner pixels / mm, scanning error ±0.05mm), acquiring the actual spatial positions of multiple measurement points, with the duct lip serving as the mold cavity end structure, synchronously included in the infrared acquisition range along with the outer contour reference surface; comparing the spatial positions of the measurement points with the designed outer contour positions of the plaster mold point by point to obtain the outer contour offset; combining the local wall thickness value at the measurement point and the outer contour offset, using a heterogeneous shrinkage field inverse distance weighted interpolation algorithm to calculate the shrinkage offset of corresponding points on the inner cavity blade surface, obtaining the current spatial position of the inner cavity blade surface; comparing the current spatial position of the inner cavity blade surface with the designed target position to generate inner cavity blade surface offset data; adjusting the clamping position of the plaster mold relative to the casting base plate according to the offset data, including the pose direction and displacement, until the inner cavity blade surface offset meets the preset tolerance; maintaining the adjusted pose of the plaster mold and performing the duct cavity mold casting process.

[0029] Among them, the integral plaster mold is a one-piece molding structure used to eliminate the parting seam of the duct blade. Its outer contour reference surface is a preset positioning reference surface, and a flat area without protrusions or depressions is selected on the outer contour of the plaster mold. The infrared scanning adopts an area array infrared scanning method, and the collected measurement points are evenly distributed on the outer contour reference plane, covering the outer contour edge, center and key corner areas, avoiding the offset calculation deviation caused by the concentration of measurement points; The heterogeneous shrinkage field inverse distance weighted interpolation algorithm is based on the uneven shrinkage characteristics caused by local wall thickness differences in plaster molds. Unlike the conventional uniform shrinkage assumption, it achieves accurate calculation of the internal blade offset through weighted interpolation.

[0030] Specifically, such as Figure 2 As shown, combining the local wall thickness and outer profile offset at the measurement point, the shrinkage offset at the corresponding point on the inner blade surface is calculated using a heterogeneous shrinkage field inverse distance weighted interpolation algorithm. The specific steps are as follows: Using the local wall thickness at the measurement point as a weighting factor, a spatial weighting function for the heterogeneous contraction field is constructed. The larger the local wall thickness at the measurement point, the smaller the contribution of the outer contour offset at that measurement point to the calculation of the contraction offset at the corresponding point on the inner blade surface. Specifically: The spatial weighting function consists of two parts: distance weights. and wall thickness weight The two parts are combined using a multiplicative fusion method to form the comprehensive weight. The functional relationship is as follows:

[0031]

[0032]

[0033] in, Points to be calculated With measurement point The three-dimensional Euclidean distance between them (unit: mm). This is the distance inverse power parameter (dimensionless, its definition is as above). For measurement points Local wall thickness at the location (the value after the aforementioned preheating correction); comprehensive weight With local wall thickness It is negatively correlated with spatial distance. of The power of the equation is negatively correlated, which reflects the difference in shrinkage characteristics of plaster molds in areas with different wall thicknesses, and the closer the measurement point, the greater the influence. In one implementation, if the local wall thickness value (after correction) of measurement point A is twice that of measurement point B, and both are at the same distance from the point P to be calculated, then the overall weight of measurement point A is half that of measurement point B, ensuring that the outer profile offset of the area with larger wall thickness does not excessively affect the calculation result of the inner cavity blade offset.

[0034] For any point to be calculated on the surface of the inner cavity blade Select the point that is spatially closest to the point to be calculated. For each measurement point, based on the inverse ratio of the spatial distance between each measurement point and the point to be estimated, and the spatial weighting function of the heterogeneous contraction field, the contraction offset of the point to be estimated is calculated using a weighted average. Specifically: The complete mathematical expression for the inverse distance weighted interpolation algorithm for heterogeneous contraction fields is:

[0035] in, Points to be calculated Shrinkage offset (unit: mm). The number of nearby measurement points selected (based on the complexity of the blade surface). ), Points to be calculated With measurement point The three-dimensional Euclidean distance between them This is the distance inverse power parameter (its definition is as before). For measurement points The local wall thickness at that location (the value after the aforementioned preheating correction). Let be the shrinkage transfer constant of the plaster mold (as defined above). For measurement points The outer contour offset at that location; Determine the points to be calculated The spatial coordinates are used as the center to define a certain spatial range, and the nearest distance within this range is selected. One measurement point (the number can be adjusted according to the complexity of the blade surface). ),calculate With each measurement point The three-dimensional Euclidean distance between them Calculate the overall weight according to the above formula. The initial value of the shrinkage offset at each measurement point. With comprehensive weight Substituting into the weighted average formula, we obtain the point to be calculated. final shrinkage offset .

[0036] In this implementation scheme, the local wall thickness value is used as the core parameter of the spatial weighting function of the heterogeneous contraction field. The weighting factor is negatively correlated with the wall thickness value. The larger the wall thickness, the weaker the response to the outer contour offset, and the smaller the wall thickness, the stronger the response. This avoids the excessive influence of the outer contour offset of the larger wall thickness region on the inner cavity calculation result. When calculating the contraction offset of any point to be calculated on the inner cavity blade surface, the adjacent measurement points are selected with the spatial coordinates of that point as the center, and the weights are allocated inversely proportional to the spatial distance. The distance weighting and the wall thickness weighting are combined for comprehensive calculation, so that the obtained contraction offset reflects the spatial distance contribution of each measurement point on the outer contour.

[0037] Specifically, the spatial location of the measurement points is compared with the designed location of the plaster mold outline point by point to obtain the outline offset. The specific steps are as follows: By performing surface fitting on the actual spatial positions of multiple measurement points, the measured point cloud surface of the plaster mold outline datum surface is obtained, specifically as follows: The least squares method was used to fit the actual spatial coordinates of all collected measurement points to a surface. Abnormal measurement points with excessive deviations during the fitting process were removed (the criterion for judging abnormal points is that the deviation from the fitted surface exceeds twice the overall measurement deviation). During the fitting process, it was ensured that the surface of the measured point cloud was consistent with the actual shape of the plaster mold outline, and key features such as corners and edges of the outline reference surface were retained. In one implementation, 100 uniformly distributed measurement points are collected, and the measured point cloud surface is obtained by fitting using the least squares method. The average deviation between the fitted surface and each measurement point is controlled within a reasonable range to ensure that the surface can truly reflect the actual state of the plaster mold outline.

[0038] The measured point cloud surface is spatially registered and aligned with the design surface of the plaster mold outline. The deviation distance of each measurement point on the measured point cloud surface along the normal direction of the design surface of the plaster mold outline is calculated to obtain the outline offset at each measurement point, which is as follows: The Iterative Closest Point Algorithm (ICP algorithm) is used to spatially register the measured point cloud surface with the outer contour design surface, so that the key feature points of the two surfaces are aligned. During the registration process, the pose of the measured point cloud surface is iteratively adjusted until the registration error meets the preset requirements. After registration, for each measurement point, along the normal direction of the outer contour design surface at that point, calculate the distance between the coordinates of the measured point cloud surface at that point and the coordinates of the design surface at that point. This distance is the outer contour offset of the measurement point. If the distance is positive, it means that the measured position deviates from the design position and bulges outward. If it is negative, it means that the measured position is concave inward. Table 1 shows an example of calculating the offset of the measurement point's outer contour. Table 1 Example of calculating the outer contour offset of the measurement point

[0039] In this implementation scheme, surface fitting is performed on the actual spatial positions of multiple measurement points. The least squares method is used to construct the measured point cloud surface of the plaster mold outline reference surface. Abnormal measurement points with excessive deviations are removed during the fitting process, thereby reducing the interference of single-point measurement noise on the overall surface shape. This allows the fitting results to accurately reproduce the key features such as corners and edges of the outline reference surface. An iterative nearest point algorithm is introduced to spatially register and align the measured point cloud surface with the design surface, so that the key feature points of the two surfaces coincide in a unified coordinate system. Subsequently, the deviation distance is calculated along the normal direction of each measurement point on the design surface. The positive or negative value of the deviation distinguishes the contraction state of the outline, which is either convex outward or concave inward, thus avoiding the projection error that may be introduced by direct coordinate difference calculation.

[0040] Specifically, by combining the local wall thickness and outer profile offset at the measurement point, a heterogeneous contraction field inverse distance weighted interpolation algorithm is used to calculate the contraction offset of the corresponding point on the inner blade surface, thereby obtaining the current spatial position of the inner blade surface. The specific steps are as follows: Based on the design models of the plaster mold's outer contour reference surface and the inner cavity blade surface, the vertical distance from the plaster mold's outer contour reference surface to the inner cavity blade surface is calculated along the normal direction at each measurement point. This distance is used as the local wall thickness value at that measurement point. Specifically: Retrieve the three-dimensional design models of the plaster mold's outer contour reference plane and the inner cavity blade surface, ensuring that the two models are in the same spatial coordinate system; For each measurement point, determine the coordinates of the point on the outer profile reference surface design model. Along the normal direction of the point (i.e. the direction perpendicular to the outer profile reference surface design surface), draw a perpendicular line from the measurement point to the inner cavity blade surface design model. The length of the perpendicular line is the local wall thickness value at the measurement point. In one embodiment, the coordinates of the measurement point N on the outer profile reference plane are (Xn, Yn, Zn), and its normal direction is the negative direction of the Z-axis. A perpendicular line is drawn along this direction, and the coordinates of the intersection point with the inner cavity blade surface design model are (Xn, Yn, Zn''). Then the local wall thickness value of the measurement point is Zn-Zn''.

[0041] The local wall thickness at each measurement point and the corresponding outer contour offset at each measurement point are input together into the heterogeneous contraction field inverse distance weighted interpolation algorithm, which is as follows: The local wall thickness, outer profile offset, and spatial coordinates of each measurement point are taken as a complete set of input parameters and synchronously input into the heterogeneous contraction field inverse distance weighted interpolation algorithm. During the input process, the parameters are standardized to avoid algorithm calculation deviations caused by differences in units; In one implementation, the local wall thickness and the outer profile offset are both converted to the same length unit. After being input into the algorithm, the algorithm automatically associates the parameters of each measurement point and, in combination with the spatial weight function and the inverse distance relationship, calculates the offset of the inner cavity blade.

[0042] In this implementation scheme, by retrieving the three-dimensional design model of the plaster mold's outer contour reference surface and the inner cavity blade surface, the vertical distance from the outer contour reference surface to the inner cavity blade surface is calculated along the normal direction of each measurement point in a unified spatial coordinate system. This vertical distance is used as the local wall thickness value at the corresponding measurement point, ensuring that the extracted wall thickness value accurately corresponds to the actual geometric structure of the plaster mold. This avoids the input deviation caused by simplifying to a uniform wall thickness. The local wall thickness value, outer contour offset, and spatial coordinates of each measurement point are simultaneously fed into the heterogeneous shrinkage field inverse distance weighted interpolation algorithm as a complete set of input parameters. The parameters are standardized before input, eliminating the algorithm calculation deviation caused by dimensional differences.

[0043] Specifically, the clamping position of the plaster mold relative to the casting base plate is adjusted according to the offset data, including the orientation and displacement, until the offset of the inner cavity blade surface meets the preset tolerance. The specific steps are as follows: The internal blade surface offset data is decomposed into translational and rotational components, specifically as follows: The spatial coordinate decomposition method is used to decompose the offset data of each feature point on the inner cavity blade surface into translational components along the X, Y, and Z coordinate axes, and rotational components around the X, Y, and Z coordinate axes. The translation component reflects the overall translational displacement of the inner cavity blade surface, while the rotation component reflects the attitude displacement of the inner cavity blade surface. In one implementation, if the inner cavity blade surface shifts as a whole in the positive X-axis direction, the translation component in the X-axis direction is positive; if it rotates clockwise around the Y-axis, the rotation component around the Y-axis is negative. By decomposing the components, the specific direction and angle that the plaster mold needs to be adjusted can be determined.

[0044] Adjusting the translational degrees of freedom of the plaster mold to eliminate the overall translational deviation of the inner cavity blade surface in three-dimensional space, specifically: Based on the translation components obtained from the decomposition, the clamping and adjustment mechanism is controlled to adjust the position of the plaster mold along the X, Y, and Z coordinate axes respectively. During the adjustment process, the offset is gradually compensated based on the translation components. In one implementation, if the translation component in the X-axis direction is ΔX, the clamping mechanism is controlled to move the plaster mold along the negative X-axis direction by ΔX. Similarly, the translation deviations in the Y-axis and Z-axis directions are adjusted respectively until the translation components in the three directions are close to zero, thus eliminating the overall translation deviation.

[0045] After eliminating the overall translational deviation, adjust the rotational degrees of freedom of the plaster mold to eliminate the angular deviation of the inner cavity blade surface around each coordinate axis. Specifically: Based on the rotational components obtained from the decomposition, the clamping and adjustment mechanism is controlled to adjust the posture angle of the plaster mold around the three coordinate axes of X, Y, and Z. During the adjustment process, a step-by-step fine-tuning method is adopted to avoid the expansion of deviation caused by a single large adjustment. In one embodiment, if the rotation component around the Z-axis is θ, the clamping mechanism is controlled to rotate the plaster mold in the opposite direction around the Z-axis by an angle of θ. After the adjustment is completed, the inner cavity blade surface offset data is collected again to check whether the angular deviation has been eliminated, until all rotation components meet the requirements.

[0046] After the adjustment is completed, the outer contour reference surface of the plaster mold is scanned again with infrared and the offset of the inner cavity blade surface is recalculated. If the offset of the inner cavity blade surface does not meet the preset tolerance, the translation adjustment and rotation adjustment are repeated.

[0047] In this implementation scheme, the offset data of the inner cavity blade surface is decomposed into translational and rotational components, which correspond to the position and orientation offsets of the plaster mold in three-dimensional space, respectively. This decouples the originally mixed offset information and clarifies the specific directions and angles required for clamping and adjustment. The translational degree of freedom is adjusted first to eliminate the overall positional deviation of the inner cavity blade surface. After the translational component approaches zero, the rotational degree of freedom is corrected, eliminating the angular deviations around each coordinate axis one by one. This avoids the mutual interference that may occur when adjusting multiple degrees of freedom at the same time. After each adjustment, an infrared scan is performed on the outer contour reference surface of the plaster mold and the offset is recalculated to verify the correction effect. If the preset tolerance is not met, the translational and rotational adjustment cycle is entered again.

[0048] Specifically, the steps for determining the preset tolerance are as follows: The inner blade surface is divided into the leading edge region, the middle section of the blade, and the trailing edge region, specifically as follows: Based on the structural characteristics of duct blades, starting from the tip of the blade, along the blade's span, the front 1 / 3 of the blade is divided into the leading edge region, the middle 1 / 3 into the middle section region, and the rear 1 / 3 into the trailing edge region. During the division process, the curvature of the blade surface is taken into account to ensure that the boundaries of each region are clear and cover the entire area of ​​the blade, avoiding region overlap or omission; Examples of regional divisions are shown in Table 2: Table 2 Example of Regional Division

[0049] A leading-edge tolerance threshold is set for the leading edge region of the blade, a mid-section tolerance threshold is set for the middle section of the blade, and a trailing-edge tolerance threshold is set for the trailing-edge region. The leading-edge tolerance threshold is less than the mid-section tolerance threshold. Specifically: Based on the aerodynamic performance requirements of the ducted blades and the structural importance of different regions, differentiated tolerance thresholds are set. The leading edge region of the blade directly affects the airflow introduction effect, and has the highest requirements for forming accuracy, therefore the tolerance threshold is the smallest. The middle section of the leaf blade is the main stress area of ​​the leaf, and the tolerance threshold is moderate. The trailing edge region of the blade affects the airflow discharge effect, and the tolerance threshold can be slightly larger than that of the leading edge region; In one implementation, the leading edge tolerance threshold is set to be less than the middle section tolerance threshold, and the trailing edge tolerance threshold can be the same as or slightly larger than the middle section tolerance threshold.

[0050] When the offset values ​​of the blade leading edge region, blade mid-section region, and blade trailing edge region in the internal blade surface offset data all fall within the corresponding tolerance threshold range, the internal blade surface offset is determined to meet the preset tolerance, which is as follows: For each feature point on the inner cavity blade surface, determine its region, and then compare the offset of the feature point with the tolerance threshold of the corresponding region. If the offset of all feature points does not exceed the tolerance threshold of the corresponding region, and the offset of more than half of the feature points in each region is close to zero, then the offset is determined to meet the preset tolerance. If the offset of a feature point in any region exceeds the corresponding tolerance threshold, or if the offset of most feature points in a certain region is large, it is determined that the preset tolerance is not met, and the plaster mold clamping position needs to be readjusted.

[0051] In this implementation scheme, the inner blade surface is divided into the leading edge region, the middle section region, and the trailing edge region according to the spanwise position and curvature characteristics. This ensures that the boundaries of each region are clear and completely covered, avoiding confusion caused by overlapping or omissions of regions. The strictest tolerance threshold is set for the leading edge region, a moderate threshold for the middle section region, and a slightly wider threshold for the trailing edge region. This aligns with the differences in airflow introduction, carrying, and exhaust functions of different parts of the ducted fan. As the part where the airflow first comes into contact and has the largest curvature, the forming accuracy of the leading edge directly determines the fan's propulsion efficiency. Therefore, the strictest passing conditions are matched. During the judgment, the offset of all regions must fall within their respective threshold ranges, and the offset of most feature points in each region must be close to zero. This prevents the out-of-tolerance of local key areas from being masked by the overall average, and avoids the hidden danger of a few points being accidentally qualified while most points deviate by judging the trend of offset distribution.

[0052] Specifically, the shrinkage offset of corresponding points on the inner blade surface is calculated using the following steps: The inner cavity blade surface is discretized into mesh nodes, specifically as follows: The structured mesh discretization method is used to divide the inner blade surface into multiple uniformly distributed mesh nodes. The mesh node density is adjusted according to the complexity of the blade surface. In areas with large curvature of the blade surface (such as the leading edge region), the mesh node density is larger, while in areas with gentle curvature of the surface (such as the middle section of the blade), the mesh node density can be appropriately reduced. In one embodiment, the inner cavity blade surface is discretized into 500-800 grid nodes, with the grid node density in the leading edge region being 1.5 times that in the middle section of the blade, ensuring that the discretized grid nodes can comprehensively and accurately reflect the shape of the inner cavity blade surface.

[0053] The measurement points corresponding to each grid node along the wall thickness direction of the plaster mold are selected on the outer contour reference surface of the plaster mold. Specifically: For each grid node, draw a perpendicular line along the wall thickness direction of the plaster mold (i.e., from the inner cavity blade surface to the outer contour reference surface). The intersection of this perpendicular line and the outer contour reference surface is the measurement point corresponding to that grid node. If the perpendicular line of a certain grid node does not intersect with the outer contour reference plane, then the measurement point of the outer contour reference plane closest to the perpendicular line is selected as the corresponding measurement point to ensure that each grid node has a unique corresponding measurement point. In one implementation, the intersection of the perpendicular line of the grid node Q along the wall thickness direction and the outer contour reference plane is the measurement point Q'. Then Q' is the measurement point corresponding to Q. Subsequently, the shrinkage offset of point Q will be calculated based on the outer contour offset and local wall thickness value of Q'.

[0054] Using the outer contour offset at the corresponding measurement point as the initial value, multiplied by the shrinkage attenuation coefficient determined by the local wall thickness, the shrinkage offset of each grid node on the inner cavity blade surface is calculated node by node. The shrinkage offset of the mold cavity body and the end lip can be calculated simultaneously, specifically as follows: Shrinkage attenuation coefficient From the measurement point Local wall thickness at the location It is determined that its functional relationship is:

[0055] in, The shrinkage transfer constant of plaster mold (unit: mm). For measurement points The local wall thickness (unit: mm) at the specified location is taken as the value obtained in the preceding steps and corrected for preheating; shrinkage attenuation coefficient. With local wall thickness There is a negative correlation; the larger the local wall thickness, the better. The smaller the value, the smaller the local wall thickness. The larger the value, the more it reflects the difference in shrinkage of the plaster mold in areas with different wall thicknesses; During calculation, the outer contour offset of the corresponding measurement point is used. Multiply by the shrinkage attenuation factor The shrinkage offset of the grid node is obtained. :

[0056] In one implementation, the calculation process is performed on each grid node one by one to obtain the shrinkage offset of all grid nodes, and then the current spatial position of the inner cavity blade surface is fitted.

[0057] In this implementation scheme, a structured mesh discretization method is used to divide the inner cavity blade surface into multiple mesh nodes. The node density is adjusted according to the curvature difference of the surface. The nodes are more densely distributed in the leading edge area with greater curvature, and the nodes can be distributed more sparsely in the middle section area with gentler curvature. This allows the discretized mesh to faithfully reproduce the geometry of the inner cavity blade surface. For each mesh node, the corresponding measurement point is traced along the wall thickness direction of the plaster mold to the outer contour reference surface to establish a spatial correspondence between the mesh node and the measurement point. When calculating, the outer contour offset of the corresponding measurement point is used as the initial value and multiplied by the shrinkage attenuation coefficient determined by the local wall thickness value. The larger the wall thickness, the smaller the attenuation coefficient, thereby reflecting the difference in shrinkage capacity of different areas of the plaster mold. After calculating node by node, the shrinkage offset of each mesh node is obtained, and then the current spatial position of the inner cavity blade surface is fitted.

[0058] Specifically, before calculating the shrinkage offset at the corresponding point on the inner blade surface by combining the local wall thickness value and the outer profile offset at the measurement point, the local wall thickness value is corrected. The specific steps are as follows: Before the plaster mold enters the mold-closing preparation state, the preheating temperature of the plaster mold is collected, specifically: Multiple temperature acquisition points are set at different locations on the plaster mold (center, edge, and corner of the outer contour reference surface). An infrared temperature measurement module is used to synchronously collect the temperature of each temperature acquisition point. The average value of the temperatures of all acquisition points is taken as the preheating temperature of the plaster mold. During the data collection process, ensure that the distance between the infrared temperature measurement module and the surface of the plaster mold is consistent to avoid temperature measurement errors caused by distance differences; In one embodiment, five temperature sampling points are set on the outer contour reference surface of the plaster mold, and the collected temperatures are T1, T2, T3, T4 and T5, respectively. The average value (T1+T2+T3+T4+T5) / 5 is taken as the preheating temperature of the plaster mold.

[0059] Based on the linear expansion coefficient of the plaster mold material and the preheating temperature, the wall thickness expansion of the plaster mold under preheating conditions is calculated as follows: Obtain the coefficient of linear expansion of plaster mold material (This coefficient is an inherent property of the plaster mold material, obtained through standard thermal expansion tests of plaster materials), and then based on the preheating temperature... With room temperature The difference Combined with the coefficient of linear expansion Calculate the expansion of the wall thickness; The calculation logic for the expansion amount is as follows: Thermal expansion wall thickness increment

[0060] in, This refers to the design wall thickness of the plaster mold. The coefficient of linear expansion of gypsum material (unit: °C) - ¹), The difference between the preheating temperature and the room temperature (unit: °C); The design wall thickness value between the design model of the outer contour datum surface of the plaster mold and the design model of the inner cavity blade surface. The increase in wall thickness due to thermal expansion The local wall thickness value after thermal expansion correction was obtained. The local wall thickness value after thermal expansion correction can reflect the actual change in wall thickness of the plaster mold due to thermal expansion under preheating conditions, ensuring the accuracy of subsequent offset calculation; this wall thickness correction logic is also adapted to the size correction of the duct lip.

[0061] Based on the water loss shrinkage characteristics of the plaster mold material and the change in moisture content during preheating, the wall thickness correction amount of the plaster mold in the preheated state is calculated, specifically as follows: Obtain the volume shrinkage coefficient of plaster mold material (This coefficient is an inherent property of plaster molding materials, obtained through a drying shrinkage test of plaster materials. The calibration method is as follows: take a standard plaster specimen and measure the ratio of its linear shrinkage rate to the change in moisture content during the process of the moisture content decreasing from the initial value to the residual moisture content.) Then, based on the average change in moisture content... Combined with the volume shrinkage coefficient of water loss Calculate the correction amount for the wall thickness; The calculation logic for the correction amount is as follows: Water loss and wall thickness shrinkage

[0062] in, This refers to the design wall thickness value of the plaster mold. This is the volume shrinkage coefficient of gypsum material (unit: mm / mm / %, i.e., the linear shrinkage rate corresponding to a 1% change in moisture content). The average value of the change in moisture content before and after preheating (unit: %). The design wall thickness value between the design model of the outer contour datum surface of the plaster mold and the design model of the inner cavity blade surface. Subtract the shrinkage due to water loss and wall thickness The local wall thickness value after water loss shrinkage correction was obtained. The local wall thickness value after water loss and shrinkage correction can reflect the actual change in wall thickness of the plaster mold due to water loss and shrinkage under preheating conditions, ensuring the accuracy of subsequent offset calculation; this wall thickness correction logic is also suitable for the size correction of the duct lip.

[0063] In this implementation scheme, before the plaster mold enters the mold-closing preparation state, multiple temperature acquisition points are set at different positions on the outer contour reference surface. The temperature of each point is collected synchronously by an infrared temperature measurement module, and the average value is taken as the preheating temperature. This avoids the local deviation that may be caused by single-point temperature measurement. Based on the linear expansion coefficient of the plaster mold material and the difference between the preheating temperature and the room temperature, the wall thickness expansion is calculated. The design wall thickness value is superimposed with this expansion to obtain the corrected local wall thickness value. The thermal expansion factor of the plaster mold in the preheating state is included in the determination of the wall thickness parameter, so that the local wall thickness value input to the algorithm is closer to the actual geometric state of the plaster mold during the mold-closing preparation. This avoids the deviation introduced by directly using the room temperature design wall thickness for subsequent interpolation calculations due to ignoring thermal expansion.

[0064] Specifically, the clamping position of the plaster mold relative to the casting base plate is adjusted based on the offset data, including the orientation and displacement. The specific steps are as follows: When the offset data of the inner cavity blade surface indicates that the entire inner cavity blade surface has shifted to one side, the offset data is converted into a reverse compensation amount for the plaster mold clamping position, specifically as follows: Determine the overall offset direction of the inner cavity blade surface. If the offset data shows that the inner cavity blade surface is offset in the positive direction of the X-axis, then the reverse compensation direction of the plaster mold clamping position is the negative direction of the X-axis, and the magnitude of the compensation amount is consistent with the magnitude of the overall offset. The conversion of compensation amount needs to combine the translation and rotation components of the offset data to ensure that the reverse compensation can accurately offset the offset of the inner cavity blades. In one embodiment, the inner cavity blade surface is offset by ΔY in the positive Y-axis direction and rotated clockwise by θ around the Z-axis. The reverse compensation amount of the plaster mold clamping position is a movement of ΔY in the negative Y-axis direction and a counterclockwise rotation by θ around the Z-axis, thereby achieving precise compensation for the offset.

[0065] The reverse compensation amount is decomposed into at least two progressive adjustment steps according to a preset ratio. After executing the first progressive adjustment step, the actual spatial position of the plaster mold outer contour reference surface is re-acquired and the inner cavity blade surface offset data is recalculated. Then, the second progressive adjustment step is executed based on the recalculated inner cavity blade surface offset data. Specifically: The reverse compensation amount is decomposed into 2-3 progressive adjustment steps, with each adjustment amount being 1 / 2 or 1 / 3 of the total compensation amount, to avoid excessive displacement of the plaster mold or the emergence of new deviations due to a single large adjustment. In one implementation, the total reverse compensation amount is Δ, which is decomposed into two adjustment steps. The first adjustment is Δ / 2. After the adjustment is completed, the outer contour reference surface of the plaster mold is scanned again with infrared to collect the actual spatial position of the measurement point and recalculate the offset data of the inner cavity blade surface. If the offset data does not meet the preset tolerance, the second adjustment Δ / 2 is performed until the offset amount meets the requirements. If there is still a deviation after the second adjustment, a small fine-tuning can be made to ensure the accuracy of the adjustment.

[0066] In this implementation scheme, when the offset data of the inner cavity blade surface indicates that the whole is offset to one side, the offset data is converted into a reverse compensation amount for the plaster mold clamping position. The compensation direction is opposite to the offset direction, and the compensation amount corresponds to the offset amount. This allows the adjustment command to be generated directly for the source of the deviation. The reverse compensation amount is decomposed into multiple progressive adjustment steps according to a preset ratio. Each step is executed by re-acquiring the actual spatial position of the outer contour reference surface and refreshing the offset calculation. The adjustment range of the next step is determined based on the latest offset data, thereby avoiding over-adjustment or local stress concentration in the mold that may be caused by a single large compensation.

[0067] Please see Figure 3 This invention provides a technical solution: an infrared-calibrated duct mold cavity optimization casting system, comprising: an infrared scanning unit for scanning the outer contour reference surface of a plaster mold with infrared, acquiring the actual spatial positions of multiple measurement points, and also for acquiring the preheating temperature of the plaster mold before it enters the mold closing preparation state; an outer contour offset acquisition unit for comparing the actual spatial positions of multiple measurement points with the designed outer contour positions of the plaster mold point by point to obtain the outer contour offset at each measurement point; and a heterogeneous shrinkage field interpolation calculation unit for calculating the wall thickness expansion of the plaster mold in the preheating state based on the linear expansion coefficient of the plaster mold material and the preheating temperature, superimposing the designed wall thickness value with the wall thickness expansion value to obtain the corrected local wall thickness value, and combining it with the correction... The local wall thickness value and the outer contour offset at each measurement point are calculated using a heterogeneous shrinkage field inverse distance weighted interpolation algorithm to estimate the shrinkage offset of the corresponding point on the inner cavity blade surface, thus obtaining the current spatial position of the inner cavity blade surface. The inner cavity offset generation unit is used to compare the current spatial position of the inner cavity blade surface with the design target position to generate inner cavity blade surface offset data. The clamping adjustment unit is used to adjust the clamping position of the plaster mold relative to the casting base plate according to the inner cavity blade surface offset data, including the pose direction and displacement, until the inner cavity blade surface offset detected after adjustment meets the preset tolerance. The casting execution unit is used to control the execution of the duct mold cavity casting operation under the adjusted pose of the plaster mold.

[0068] The units interact bidirectionally via a data transmission module. The infrared scanning unit includes an infrared 3D scanner, an infrared temperature measurement module, and a data acquisition module. The infrared 3D scanner is used to achieve full-area scanning of the outer contour reference surface of the plaster mold. The infrared temperature measurement module is used to collect the preheating temperature of the plaster mold before it enters the mold closing preparation state. The data acquisition module is used to simultaneously collect the spatial coordinate data and preheating temperature data of the measurement points, and to perform noise reduction processing on the spatial coordinate data to remove abnormal data points. The outer contour offset acquisition unit has a built-in surface fitting module and a spatial registration module. The surface fitting module is used to fit the coordinates of the measurement points to obtain the measured point cloud surface. The spatial registration module is used to register and align the measured point cloud surface with the design surface and calculate the outer contour offset. The heterogeneous contraction field interpolation calculation unit has a built-in algorithm calculation module, wall thickness correction module and grid discretization module. The wall thickness correction module is used to correct the local wall thickness value for preheating expansion. The grid discretization module is used to discretize the inner cavity blade surface into grid nodes. The algorithm calculation module is used to execute the heterogeneous contraction field inverse distance weighted interpolation algorithm to complete the calculation of the inner cavity blade offset. The internal cavity offset generation unit has a built-in deviation comparison module, which is used to compare the current spatial position of the internal cavity blade with the design target position point by point, generate offset data, and mark the feature points where the offset exceeds the standard. The clamping and adjustment unit includes a displacement adjustment module, an attitude adjustment module, and a closed-loop control module. The displacement adjustment module is used to adjust the translation position of the plaster mold, the attitude adjustment module is used to adjust the orientation of the plaster mold, and the closed-loop control module is used to realize the closed-loop control of adjustment-detection-re-adjustment to ensure that the offset meets the preset tolerance. The casting execution unit includes a preheating module, a pouring module, and a demolding module. The preheating module is used to preheat the plaster mold, the pouring module is used to control the injection process of the casting material, and the demolding module is used to remove the plaster mold and take out the casting after the casting material has solidified. The working sequence of each unit is uniformly controlled by the central control unit. The central control unit receives feedback data from each unit, coordinates the work of each unit synchronously, ensures the stable operation of the entire casting system, and realizes the precise casting of the duct cavity.

[0069] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0070] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for optimizing duct cavity casting based on infrared calibration, characterized in that, Includes the following steps: Infrared scanning was performed on the outer contour reference surface of the plaster mold to collect the actual spatial positions of multiple measurement points; The actual spatial positions of multiple measurement points are compared with the designed positions of the plaster mold outline point by point to obtain the outline offset at each measurement point; By combining the local wall thickness and outer profile offset at each measurement point, the inverse distance weighted interpolation algorithm of the heterogeneous contraction field is used to calculate the contraction offset of the corresponding point on the inner cavity blade surface and obtain the current spatial position of the inner cavity blade surface. The current spatial position of the inner cavity blade surface is compared with the design target position to generate inner cavity blade surface offset data. Adjust the clamping position of the plaster mold relative to the casting base plate based on the offset data of the inner cavity blade surface, including the orientation and displacement, until the offset of the inner cavity blade surface detected after adjustment meets the preset tolerance. Maintain the adjusted position of the plaster mold and perform the duct cavity casting process.

2. The duct cavity optimization casting method based on infrared calibration according to claim 1, characterized in that, By combining the local wall thickness and outer profile offset at each measurement point, the shrinkage offset at corresponding points on the inner blade surface is calculated using a heterogeneous shrinkage field inverse distance weighted interpolation algorithm. The specific steps are as follows: A spatial weighting function for the heterogeneous contraction field is constructed using the local wall thickness at each measurement point as a weighting factor. For any point to be calculated on the curved surface of the inner blade, select multiple measurement points that are spatially closest to the point to be calculated. Based on the inverse ratio of the spatial distance between each measurement point and the point to be calculated, and the spatial weighting function of the heterogeneous contraction field, calculate the contraction offset of the point to be calculated in a weighted manner.

3. The duct cavity optimization casting method based on infrared calibration according to claim 1, characterized in that, The actual spatial positions of multiple measurement points are compared with the designed positions of the plaster mold outline point by point to obtain the outline offset at each measurement point. The specific steps are as follows: By performing surface fitting on the actual spatial positions of multiple measurement points, the measured point cloud surface of the plaster mold outline reference surface is obtained; The measured point cloud surface is spatially registered and aligned with the plaster mold outline design surface. The deviation distance of each measurement point on the measured point cloud surface along the normal direction of the plaster mold outline design surface is calculated to obtain the outline offset at each measurement point.

4. The duct cavity optimization casting method based on infrared calibration according to claim 2, characterized in that, Obtain the local wall thickness values ​​at each measurement point, including: Based on the design model of the plaster mold outer contour reference surface and the design model of the inner cavity blade surface, the vertical distance from the plaster mold outer contour reference surface to the inner cavity blade surface is calculated along the normal direction of each measurement point, and used as the local wall thickness value at each measurement point. The local wall thickness and outer profile offset at each measurement point are used to input the heterogeneous contraction field inverse distance weighted interpolation algorithm.

5. The duct cavity optimization casting method based on infrared calibration according to claim 1, characterized in that, Adjust the clamping position of the plaster mold relative to the casting base plate based on the offset data of the inner cavity blade surface, including the orientation and displacement, until the offset of the inner cavity blade surface detected after adjustment meets the preset tolerance. The specific steps are as follows: The internal blade surface offset data is decomposed into translational and rotational components. Adjust the translational degrees of freedom of the plaster mold; After eliminating the overall translational deviation, adjust the rotational degrees of freedom of the plaster mold; After the adjustment is completed, the outer contour reference surface of the plaster mold is scanned again with infrared and the offset of the inner cavity blade surface is recalculated. If the offset of the inner cavity blade surface does not meet the preset tolerance, the translation adjustment and rotation adjustment are repeated.

6. The duct cavity optimization casting method based on infrared calibration according to claim 5, characterized in that, Determining whether the offset of the inner cavity blade surface meets the preset tolerance includes: The inner blade surface is divided into the leading edge area, the middle section of the blade, and the trailing edge area. Set a leading edge tolerance threshold for the leading edge region of the blade, a middle section tolerance threshold for the middle section of the blade, and a trailing edge tolerance threshold for the trailing edge region of the blade. The leading edge tolerance threshold is less than the middle section tolerance threshold. When the offset values ​​of the blade leading edge region, blade mid-section region, and blade trailing edge region in the internal blade surface offset data all fall within the corresponding tolerance threshold range, it is determined that the internal blade surface offset value meets the preset tolerance.

7. The duct cavity optimization casting method based on infrared calibration according to claim 1, characterized in that, The specific steps for calculating the shrinkage offset of corresponding points on the curved surface of the inner blade are as follows: Discretize the inner cavity blade surface into mesh nodes; Select measurement points on the outer contour reference surface of the plaster mold that correspond to each grid node along the wall thickness direction of the plaster mold; Using the outer profile offset at the corresponding measurement point as the initial value, multiply it by the shrinkage attenuation coefficient determined by the local wall thickness value, and calculate the shrinkage offset of each grid node on the inner cavity blade surface node by node.

8. The duct cavity optimization casting method based on infrared calibration according to claim 1, characterized in that, Before calculating the shrinkage offset at corresponding points on the inner blade surface by combining the local wall thickness and outer profile offset at each measurement point and using the heterogeneous shrinkage field inverse distance weighted interpolation algorithm, the local wall thickness is corrected. The specific steps are as follows: Before the plaster mold enters the mold-closing preparation state, the preheating temperature of the plaster mold is collected; Calculate the wall thickness expansion of the plaster mold under preheating conditions based on the linear expansion coefficient of the plaster mold material and the preheating temperature. The design wall thickness value between the design model of the outer contour datum surface of the plaster mold and the design model of the inner cavity blade surface is superimposed with the wall thickness expansion amount to obtain the corrected local wall thickness value.

9. The duct cavity optimization casting method based on infrared calibration according to claim 1, characterized in that, Adjust the clamping position of the plaster mold relative to the casting base plate based on the offset data of the inner cavity blade surface, including the orientation and displacement. The specific steps are as follows: When the internal blade surface offset data indicates that the internal blade surface is offset to one side, the internal blade surface offset data is converted into the reverse compensation amount of the plaster mold clamping position. The reverse compensation amount is decomposed into at least two progressive adjustment steps according to a preset ratio. After the first progressive adjustment step is executed, the actual spatial position of the plaster mold outer contour reference surface is re-acquired and the inner cavity blade surface offset data is recalculated. Then, the second progressive adjustment step is executed based on the recalculated inner cavity blade surface offset data.

10. A duct cavity optimization casting system based on infrared calibration, employing the duct cavity optimization casting method based on infrared calibration according to any one of claims 1-9, characterized in that, include: The infrared scanning unit is used to perform infrared scanning on the outer contour reference surface of the plaster mold, collect the actual spatial position of multiple measurement points, and also to collect the preheating temperature of the plaster mold before it enters the mold closing preparation state. The outer contour offset acquisition unit is used to compare the actual spatial position of multiple measurement points with the design position of the plaster mold outer contour point by point to obtain the outer contour offset at each measurement point. The heterogeneous shrinkage field interpolation calculation unit is used to calculate the wall thickness expansion of the plaster mold in the preheated state based on the linear expansion coefficient of the plaster mold material and the preheating temperature. The design wall thickness value is superimposed with the wall thickness expansion value to obtain the corrected local wall thickness value. Combined with the corrected local wall thickness value and the outer contour offset at each measurement point, the heterogeneous shrinkage field inverse distance weighted interpolation algorithm is used to calculate the shrinkage offset of the corresponding point on the inner cavity blade surface and obtain the current spatial position of the inner cavity blade surface. The internal cavity offset generation unit is used to compare the current spatial position of the internal cavity blade surface with the design target position to generate internal cavity blade surface offset data. The clamping adjustment unit is used to adjust the clamping position of the plaster mold relative to the casting base plate according to the offset data of the inner cavity blade surface, including the orientation and displacement, until the offset of the inner cavity blade surface detected after adjustment meets the preset tolerance. The casting execution unit is used to control the casting and forming operation of the duct cavity under the adjusted position of the plaster mold.

Images